Bootstrap confidence sets under model misspecification

DOI10.1214/15-AOS1355zbMATH Open1327.62179arXiv1410.0347MaRDI QIDQ892253FDOQ892253

Authors: V. G. Spokoiny, M. Zhilova 

Publication date: 18 November 2015

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: A multiplier bootstrap procedure for construction of likelihood-based confidence sets is considered for finite samples and a possible model misspecification. Theoretical results justify the bootstrap validity for a small or moderate sample size and allow to control the impact of the parameter dimension $p$: the bootstrap approximation works if $p^3/n$ is small. The main result about bootstrap validity continues to apply even if the underlying parametric model is misspecified under the so-called small modelling bias condition. In the case when the true model deviates significantly from the considered parametric family, the bootstrap procedure is still applicable but it becomes a bit conservative: the size of the constructed confidence sets is increased by the modelling bias. We illustrate the results with numerical examples for misspecified linear and logistic regressions.

Full work available at URL: https://arxiv.org/abs/1410.0347

Recommendations

• Bootstrap inference for misspecified moment condition models
• scientific article; zbMATH DE number 1794221
• scientific article; zbMATH DE number 850142
• Balanced Simultaneous Confidence Sets
• A recentred bootstrap procedure for constructing uniformly correct confidence sets under smooth function models

zbMATH Keywords

Gaussian approximationPinsker's inequalityfinite sample sizelikelihood-based bootstrap confidence setmultiplier/weighted bootstrap

Mathematics Subject Classification ID

Bootstrap, jackknife and other resampling methods (62F40) Parametric tolerance and confidence regions (62F25) Approximations to statistical distributions (nonasymptotic) (62E17)

Cites Work

• Weak convergence and empirical processes. With applications to statistics
• Regression Quantiles
• Bootstrap methods: another look at the jackknife
• Title not available (Why is that?)
• Bootstrap procedures under some non-i.i.d. models
• When does bootstrap work! Asymptotic results and simulations
• Jackknife, bootstrap and other resampling methods in regression analysis
• Bootstrap and wild bootstrap for high dimensional linear models
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• The Large-Sample Distribution of the Likelihood Ratio for Testing Composite Hypotheses
• The bootstrap and Edgeworth expansion
• The weighted bootstrap
• Title not available (Why is that?)
• Title not available (Why is that?)
• Some nonasymptotic results on resampling in high dimension. I: Confidence regions
• How do bootstrap and permutation tests work?
• Parametric estimation. Finite sample theory
• Weighted bootstrapping of \(U\)-statistics
• Generalized bootstrap for estimating equations
• Efficient estimation of semiparametric conditional moment models with possibly nonsmooth residuals
• Multiplier bootstrap of tail copulas with applications
• Robust semiparametric M-estimation and the weighted bootstrap
• Higher order properties of the wild bootstrap under misspecification
• Bootstrap confidence sets under model misspecification
• Smooth minimum distance estimation and testing with conditional estimating equations: uniform in bandwidth theory
• Bootstrap tests for misspecified models, with application to clustered binary data.

Cited In (22)

• Nonclassical Berry-Esseen inequalities and accuracy of the bootstrap
• Bootstrap consistency for quadratic forms of sample averages with increasing dimension
• Confidence limits to the distance of the true distribution from a misspecified family by bootstrap
• Parametric bootstrap under model mis-specification
• Bootstrap confidence sets under model misspecification
• A new perspective on robust \(M\)-estimation: finite sample theory and applications to dependence-adjusted multiple testing
• Elements of statistical inference in 2-Wasserstein space
• Multiscale change point detection
• On proximity of distributions of successive sums with respect to the Prokhorov distance
• Gaussian differentially private robust mean estimation and inference
• Large-dimensional central limit theorem with fourth-moment error bounds on convex sets and balls
• Bounding the difference between true and nominal rejection probabilities in tests of hypotheses about instrumental variables models
• Inference in a Class of Optimization Problems: Confidence Regions and Finite Sample Bounds on Errors in Coverage Probabilities
• Robust inference via multiplier bootstrap
• Large ball probabilities, Gaussian comparison and anti-concentration
• Refining Bootstrap Simultaneous Confidence Sets
• Bootstrap confidence sets for spectral projectors of sample covariance
• Bootstrap confidence regions based on M-estimators under nonstandard conditions
• Specification tests in semiparametric transformation models -- a multiplier bootstrap approach
• Nonasymptotic estimates for the closeness of Gaussian measures on balls
• Confidence sets for spectral projectors of covariance matrices
• Nonasymptotic one- and two-sample tests in high dimension with unknown covariance structure

This page was built for publication: Bootstrap confidence sets under model misspecification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q892253)